magnitude

7.4k

Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, Intel, or just a CPU.

github.com/magnitudedev/magnitude

Install magnitude with one command

Fresh Ubuntu/Debian server, 2 GB RAM, 1 CPU at least.

Success rate100.0%

Deploys1

Errors0

Warnings1

Deploys per week

Week 44

Based on 2 deploys recorded in the last 1 week. The totals above cover the last 365 days.

Known issues & fixes

ProblemWhereCountStatus

No results match your search.